Robust adaptive beamforming for impulse noise environments

One of the main problems facing beamforming in array system is impulse noise. Although traditional Minimum Variance Distortionless Response (MVDR) beamformer [1] based on classical techniques under minimizing a least-squares objective function has the optimum capabilities of signal protection, interference removal and noise reduction in Gaussian noise environment, it loses optimality when the noise such as impulse noise deviates from Gaussian distribution. In this paper, an effective beamforming algorithm based on a robust objective function is proposed to mitigate impulse noise. The proposed method does not require the prior knowledge of impulse noise distribution and can give a closed-form solution. A comparison is performed between the proposed algorithm and MVDR. The proposed algorithm clearly outperforms MVDR.